Sanjana Govil

From Legal Transactions to the Innovation Side of the Table

Govil began her career advising social entrepreneurs on legal transactions. The work was solid, the clients were fascinating, and after a while the imbalance became impossible to ignore. What her clients were building, she decided, was considerably more interesting than the legal machinery humming in the background. She wanted to be on the innovation side of the table, and she wanted that innovation aimed at public benefit.

So she made the transition deliberately rather than hoping for a lucky break. An MBA at Oxford served as the bridge into strategy consulting and research in the social impact sector, and that work handed her something she had not expected: sustained exposure to how data actually moves, or fails to move, through the social and public sectors.

What she found was an abundance nobody was using. Frontline nonprofits collect survey data constantly, asking the communities they serve what they need and what they think of the services they have received. Then the data sits there, because trawling through it and extracting rich insight takes resources that grassroots organizations rarely have. Meanwhile, at the other end of the system, philanthropic money flows downward. Funders decide strategically where capital should go, and the community voice that could sharpen those decisions almost never bubbles up to the funder level. Govil saw a loop that had been left open.

Unseen Insight and a Deep End Worth Diving Into

Her response was to build. This was before ChatGPT and Claude were in anyone's pocket, and the question she was asking was a genuinely hard one: could natural language processing draw usable insight out of vast quantities of unstructured, qualitative community feedback, and could those insights then be handed to philanthropic foundations to inform where the money actually went?

That idea became Unseen Insight, the startup Govil founded and led as CEO. She secured funding from Innovate UK to run a feasibility study, assembled a team and built an MVP, something real that could be tested against the world rather than argued about in a deck. Raising the capital to scale the business proved difficult, and Govil is matter of fact about it. What she carried out of the experience was more durable than a cap table: a working command of NLP models and a concrete sense of what it means to apply AI to messy human data. It also gave her a team to build and hire, her first real exercise in assessing talent for a problem nobody had solved before.

The People Who Took the Leap

Ask Govil about the people who shaped her and she does not reach for a famous name. She reaches for the colleagues who bet on an unproven idea. Sal Kimmich, who joined Unseen Insight as lead NLP engineer, is the first she names. Kimmich built the MVP and, just as importantly, absorbed a founder's stream of what Govil cheerfully calls extremely basic questions about what AI is and what it can do. "Sal is one of the smartest people that I've ever worked with," she says, crediting that patience and deep technical expertise with the entire learning curve that followed.

Her co-founder Tony Chan she describes as the operational heartbeat of the startup, the reason the machinery ran as well as it did. Pragyan was on the front lines, keeping things moving day to day. What unites all three in Govil's telling is not a skill set but a decision: each of them took a leap of faith to join a venture solving a distinctly niche problem.

The same generosity, she notes, made her current role possible. Oliver Southwick and Tom Rintoul were among those at Social Finance who backed her transition into a technical field. She is clear eyed about how that could have gone. Someone could easily have looked at a lawyer and researcher and asked what she knew about AI, data and digital. Instead they saw the potential and gave her the chance. Govil now hires for her own team, and that experience of being backed on trajectory rather than credentials sits close to the surface when she does.

Why This Moment in the Public Sector Excites Her

Govil's read on AI is refreshingly unromantic about the technology and genuinely optimistic about the institutions. The capability that matters most to her is AI's ability to transform pathways to access services. She points to the NHS and its 10 year plan, so much of which hinges on improving how quickly people can reach good healthcare, and to the real spotlight now trained on what AI could contribute there. Technology that widens access to crucial rights and services is, in her view, the version worth building.

But the shift she finds most exciting is cultural rather than computational. Public service organizations are open to innovation in a way they simply were not before, and she credits generative AI for that. Because Claude and ChatGPT are in everyone's pocket, because ordinary people have played with them and seen what they can do, leaders across the public sector are now willing to have the conversation. How could I be using technology to my advantage? How could I be using AI?

She is quick to add the practitioner's caveat. A general purpose chatbot will often not be the right solution for the problem in front of them. What matters is that the door is open. "The fact that they actually want to sit and reimagine all of these processes now," she says, and that they carry so much optimism about what technology could do for their domains, is the opening she has been waiting for since she first noticed all that unread community feedback. The work now is to make sure the answers those organizations get are worth the trust they are extending.